The efficacy of steroid therapy in sarcoidosis is questionable and has been controversial. Some workers have reported favorable results while others were able to achieve only temporary improvement without averting the serious sequalae of sarcoidosis such as fibrosis and cystic formation of the lungs. Because of the long spontaneous remissions characteristic of sarcoidosis, it has always been difficult to ascribe the results obtained to therapy. The present investigation is intended to evaluate the long term effects of corticosteroid therapy in sarcoidosis. The proposed controlled double blind study would include a relatively large sample of sarcoidosis patients who will be followed for four to five year periods. Provisions are made to reduce the experimental error by using rigid criteria in diagnosis and evaluation of therapy and through stratification and the double blind approach.